arxiv: 2605.01884 · v1 · submitted 2026-05-03 · 📡 eess.IV

Recognition: unknown

Cardiac Mesh Flow: One-Step Generation of 3D+t Cardiac Four-Chamber Meshes via Flow Matching

Qiang Ma , Qingjie Meng , Mengyun Qiao , Paul M. Matthews , Declan P. O'Regan , Wenjia Bai

Authors on Pith no claims yet

Pith reviewed 2026-05-09 15:47 UTC · model grok-4.3

classification 📡 eess.IV

keywords cardiac mesh generationflow matching3D+t modelingdeformation fieldsanatomical correspondenceconditional synthesisfour-chamber heart

0 comments

The pith

Cardiac Mesh Flow generates 3D+t four-chamber heart meshes in one step by warping a template with learned deformation fields.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces Cardiac Mesh Flow to generate spatio-temporal cardiac meshes that preserve anatomical correspondence and temporal coherence. Existing methods either use volumetric representations without correspondence or VAEs with fidelity-diversity trade-offs. By using flow matching to learn multi-scale deformation fields, it warps a single template to create diverse, consistent meshes across the cardiac cycle. This matters because it enables better understanding of heart structure and function at population scale through controllable synthetic data. The model also supports conditioning on chamber volumes for precise control.

Core claim

Cardiac Mesh Flow is a novel generative flow model for 3D+t cardiac four-chamber mesh generation. It leverages flow matching to perform efficient one-step generation of multi-scale free-form deformation fields that warp a template mesh, ensuring anatomical correspondence, temporal coherence, and periodic consistency. It enables controllable generation conditioned on cardiac chamber volumes and demonstrates high fidelity and diversity in experiments.

What carries the argument

Multi-scale free-form deformation fields learned via flow matching, which warp a single template mesh to generate the 3D+t meshes.

Load-bearing premise

That deforming one fixed template mesh with learned multi-scale free-form deformation fields can produce anatomically valid four-chamber meshes maintaining correspondence and coherence for many different subjects and full cardiac cycles.

What would settle it

Observing whether generated meshes exhibit non-physiological features like chamber intersections or loss of temporal smoothness when compared to real patient data sequences.

Figures

Figures reproduced from arXiv: 2605.01884 by Declan P. O'Regan, Mengyun Qiao, Paul M. Matthews, Qiang Ma, Qingjie Meng, Wenjia Bai.

**Figure 1.** Figure 1: Comparison between CardiacFlow (Ma et al., 2025) and Cardiac Mesh Flow. Left: CardiacFlow uses a one-step generative flow to learn a latent distribution, which is then decoded into a four-chamber segmentation map. Middle: Cardiac Mesh Flow employs a fusion network to predict multi-scale initial values as the inputs of a flow U-Net to generate multi-scale FFD fields. The FFD fields warp a template mesh into… view at source ↗

**Figure 2.** Figure 2: The architecture of HeartFFDNet. HeartFFDNet learns to predict multi-scale free-form deformation (FFD) fields from an input 3D cardiac four-chamber segmentation map. The corresponding mesh is reconstructed by warping a template mesh according to the predicted FFD fields. The multi-scale FFD fields predicted by HeartFFDNet serve as the training data for Cardiac Mesh Flow. After training, we can generate new… view at source ↗

**Figure 3.** Figure 3: The architecture of multi-scale flow U-Net. The flow U-Net consists of 3D convolutional blocks following adaptive instance normalisation (AdaIN) layers, which capture the features of the integration time t and conditioning variable c. The size of the multi-scale inputs and outputs correspond to the multi-scale FFD fields required for cardiac four-chamber mesh reconstruction. learnable embedding periodic Ga… view at source ↗

**Figure 4.** Figure 4: The architecture of multi-scale fusion network. A learnable embedding and a periodic Gaussian kernel encoding of time frame are fused to predict multi-scale initial values for the flow ODE system. inference, we randomly sample the embedding ˆe from an empirical Gaussian distribution N(µe, Σe), which is derived from the learnable embeddings {eϕ} of all training samples. Periodic Gaussian kernel encoding. S… view at source ↗

**Figure 5.** Figure 5: Examples of 3D+t cardiac four-chamber meshes generated by Cardiac Mesh Flow. we set σ = 1 for periodic Gaussian Kernel encoding Kσ. The number of resolution scales is set to L = 3. All experiments are conducted on an Nvidia RTX 3090 GPU with 24GB memory. 3.2. Unconditional generation Evaluation metrics. For unconditional generation of 3D+t cardiac meshes, we run Cardiac Mesh Flow and baseline methods to g… view at source ↗

**Figure 6.** Figure 6: Conditional generation of 3D+t cardiac four-chamber meshes using different conditioning variables. The different conditions are highlighted in red colour. 3.3. Conditional generation Evaluation metrics. For conditional generation, we use cardiac four-chamber phenotypes (Bai et al., 2020) as input conditioning variables, including LV myocardial mass (LVM), ventricular end-diastolic and end-systolic volume… view at source ↗

read the original abstract

Spatio-temporal (3D+t) generative modelling of cardiac shape and motion is crucial for understanding heart structure and function at population scale. Existing generative models for cardiac shape synthesis either adopt volumetric shape representations that lack anatomical correspondence across different time points and subjects, or rely on VAE-based frameworks that suffer from a trade-off between reconstruction fidelity and generative diversity. In this work, we propose Cardiac Mesh Flow, a novel generative flow model for 3D+t cardiac four-chamber mesh generation with anatomical correspondence, temporal coherence, and periodic consistency. Leveraging the flow matching technique, Cardiac Mesh Flow performs efficient one-step generation of multi-scale free-form deformation fields, which warp a template mesh to generate cardiac four-chamber meshes across a cardiac cycle. Furthermore, Cardiac Mesh Flow enables controllable generation conditioned on cardiac chamber volumes, allowing precise control of the synthetic heart. Experimental results demonstrate that Cardiac Mesh Flow achieves high fidelity and diversity on both unconditional and conditional generation, compared to state-of-the-art 3D+t cardiac mesh generation methods.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper introduces flow matching to generate 3D+t four-chamber cardiac meshes by warping a single template with multi-scale deformation fields and volume conditioning, but the abstract supplies no numbers or validation details to support the fidelity claims.

read the letter

The core contribution is a flow-matching model that produces one-step multi-scale free-form deformation fields to warp a fixed template mesh into time-series four-chamber cardiac shapes, with explicit conditioning on chamber volumes. This setup aims to deliver anatomical correspondence, temporal coherence, and periodic consistency while allowing controllable generation. The framing against prior VAE and volumetric methods is clear, and the choice of mesh representation plus flow matching for efficiency is a reasonable technical step forward for this subfield. Volume conditioning is a practical addition for downstream uses like population modeling or synthetic data creation. The abstract positions the work as outperforming existing 3D+t mesh generators on fidelity and diversity, which is the main claim to evaluate. The single-template warping approach is the part that needs the most scrutiny. Large inter-subject variation in cardiac anatomy means the learned deformations must avoid topology violations, self-intersections, or loss of chamber connectivity without explicit diffeomorphic constraints or post-hoc validity checks. The abstract offers no quantitative metrics, error bars, dataset descriptions, or ablation results, so it is impossible to tell whether those issues are actually controlled. If the full paper includes such checks and shows they hold across diverse subjects, the method gains credibility; otherwise the central assumption remains untested. This work is mainly for researchers building generative models for cardiac shape and motion or needing large sets of consistent 3D+t meshes. Readers focused on mesh-based synthesis with temporal structure and conditioning would find the approach worth examining, provided the results section supplies the missing numbers. The idea is grounded enough and the application area important enough that it should go to peer review rather than desk rejection, even if the current draft needs added evidence and validation experiments.

Referee Report

2 major / 2 minor

Summary. The paper proposes Cardiac Mesh Flow, a flow-matching generative model for one-step synthesis of 3D+t four-chamber cardiac meshes. It learns multi-scale free-form deformation fields that warp a single template mesh to produce outputs with anatomical correspondence, temporal coherence, and periodic consistency. The model supports both unconditional generation and conditional generation controlled by cardiac chamber volumes, and claims to achieve higher fidelity and diversity than prior SOTA 3D+t cardiac mesh methods.

Significance. If the central claims hold, the work would offer a meaningful advance in spatio-temporal cardiac shape modeling by providing an efficient, correspondence-preserving alternative to volumetric or VAE-based generators. The one-step flow-matching formulation and volume-conditioned control could enable scalable population-level synthesis for simulation and analysis, provided the generated meshes remain anatomically valid across diverse subjects.

major comments (2)

[Abstract] Abstract: performance claims of 'high fidelity and diversity' are asserted without any quantitative metrics, error bars, dataset details, ablation studies, or statistical comparisons; this absence makes it impossible to assess whether the data support the central claims of superiority over SOTA methods.
[Methods] Methods (deformation field generation): the core mechanism of warping a single fixed template via learned multi-scale free-form deformation fields lacks explicit topology-preservation, diffeomorphic, or self-intersection constraints; without such safeguards or quantitative validation (e.g., intersection rates, chamber connectivity checks), the approach risks producing invalid meshes for subjects whose anatomy deviates substantially from the template, directly undermining the fidelity and anatomical-validity assertions.

minor comments (2)

[Abstract] Abstract: the single long paragraph is dense; splitting the description of the method, conditioning, and results would improve readability.
[Abstract] Notation: the terms 'multi-scale free-form deformation fields' and 'periodic consistency' are introduced without immediate formal definition or reference to the precise mathematical formulation used.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback on our manuscript. We address each major comment below and outline the revisions we will make to strengthen the paper.

read point-by-point responses

Referee: [Abstract] Abstract: performance claims of 'high fidelity and diversity' are asserted without any quantitative metrics, error bars, dataset details, ablation studies, or statistical comparisons; this absence makes it impossible to assess whether the data support the central claims of superiority over SOTA methods.

Authors: We acknowledge that the abstract, as a concise summary, does not include specific quantitative metrics or dataset details. The full manuscript provides these in the Experiments section, including fidelity and diversity metrics with comparisons to prior SOTA methods, along with ablation studies and statistical analysis. To address the concern directly, we will revise the abstract to incorporate key quantitative results (e.g., primary fidelity/diversity scores and SOTA comparisons) while respecting length constraints. revision: yes
Referee: [Methods] Methods (deformation field generation): the core mechanism of warping a single fixed template via learned multi-scale free-form deformation fields lacks explicit topology-preservation, diffeomorphic, or self-intersection constraints; without such safeguards or quantitative validation (e.g., intersection rates, chamber connectivity checks), the approach risks producing invalid meshes for subjects whose anatomy deviates substantially from the template, directly undermining the fidelity and anatomical-validity assertions.

Authors: We agree that the current formulation does not impose explicit topology-preservation, diffeomorphic, or self-intersection constraints on the learned deformation fields. The flow-matching training on real cardiac meshes encourages anatomical plausibility and correspondence, but we recognize the value of additional safeguards. In the revised manuscript, we will add quantitative validation metrics such as self-intersection rates and chamber connectivity checks, and we will discuss limitations for anatomies that deviate substantially from the template. revision: yes

Circularity Check

0 steps flagged

No circularity: new generative architecture with independent experimental validation

full rationale

The paper's derivation introduces Cardiac Mesh Flow as a flow-matching model that generates multi-scale free-form deformation fields to warp a fixed template mesh, producing 3D+t four-chamber meshes. This rests on the established flow-matching framework (external technique) plus standard mesh-warping operations, with claims of fidelity and diversity backed by direct comparisons to prior SOTA methods rather than any self-referential fitting, redefinition of inputs as outputs, or load-bearing self-citations. No equations or steps reduce the generated meshes or conditioning on volumes to the model's own fitted parameters by construction; the central claims remain independent of the inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides no explicit free parameters, axioms, or invented entities beyond the high-level description of the flow model and template mesh; full paper would be needed to audit these.

pith-pipeline@v0.9.0 · 5498 in / 1141 out tokens · 32848 ms · 2026-05-09T15:47:46.165869+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

139 extracted references · 2 canonical work pages · 1 internal anchor

[1]

Medical Image Analysis , volume=

L. Medical Image Analysis , volume=
[2]

Experimental Physiology , volume=

Computational cardiac atlases: from patient to population and back , author=. Experimental Physiology , volume=
[3]

Young, Alistair A and Cowan, Brett R and Thrupp, Steven F and Hedley, Warren J and Dell’Italia, Louis J , journal=
[4]

IEEE Transactions on Medical Imaging , volume=

Hoogendoorn, Corn. IEEE Transactions on Medical Imaging , volume=
[5]

PLoS Computational Biology , volume=

Linking statistical shape models and simulated function in the healthy adult human heart , author=. PLoS Computational Biology , volume=
[6]

Puyol-Anton, Esther and Sinclair, Matthew and Gerber, Bernhard and Amzulescu, Mihaela Silvia and Langet, Helene and De Craene, Mathieu and Aljabar, Paul and Piro, Paolo and King, Andrew P , journal=
[7]

Frontiers in Cardiovascular Medicine , volume=

Machine learning approaches for myocardial motion and deformation analysis , author=. Frontiers in Cardiovascular Medicine , volume=
[8]

Nature Biomedical Engineering , volume=

Computationally guided personalized targeted ablation of persistent atrial fibrillation , author=. Nature Biomedical Engineering , volume=. 2019 , publisher=

2019
[9]

Nature Biomedical Engineering , volume=

Personalized virtual-heart technology for guiding the ablation of infarct-related ventricular tachycardia , author=. Nature Biomedical Engineering , volume=. 2018 , publisher=

2018
[10]

Circulation Research , volume=

Whole-heart modeling: applications to cardiac electrophysiology and electromechanics , author=. Circulation Research , volume=
[11]

Circulation: Arrhythmia and Electrophysiology , volume=

Predicting atrial fibrillation recurrence by combining population data and virtual cohorts of patient-specific left atrial models , author=. Circulation: Arrhythmia and Electrophysiology , volume=
[12]

Multi-class point cloud completion networks for

Beetz, Marcel and Banerjee, Abhirup and Ossenberg-Engels, Julius and Grau, Vicente , journal=. Multi-class point cloud completion networks for
[13]

IEEE Reviews in Biomedical Engineering , year=

Solving the inverse problem of electrocardiography for cardiac digital twins: A survey , author=. IEEE Reviews in Biomedical Engineering , year=
[14]

2023 , publisher=

Meng, Qingjie and Bai, Wenjia and O’Regan, Declan P and Rueckert, Daniel , journal=. 2023 , publisher=

2023
[15]

Multi-view hybrid graph convolutional network for volume-to-mesh reconstruction in cardiovascular

Gaggion, Nicol. Multi-view hybrid graph convolutional network for volume-to-mesh reconstruction in cardiovascular. Medical Image Analysis , pages=. 2025 , publisher=

2025
[16]

Correction of slice misalignment in multi-breath-hold cardiac

Villard, Benjamin and Zacur, Ernesto and Dall’Armellina, Erica and Grau, Vicente , booktitle=. Correction of slice misalignment in multi-breath-hold cardiac
[17]

International Workshop on Statistical Atlases and Computational Models of the Heart , pages=

Xu, Yiyang and Xu, Hao and Sinclair, Matthew and Puyol-Ant. International Workshop on Statistical Atlases and Computational Models of the Heart , pages=
[18]

2024 , publisher=

Muffoletto, Marica and Xu, Hao and Burns, Richard and Suinesiaputra, Avan and Nasopoulou, Anastasia and Kunze, Karl P and Neji, Radhouene and Petersen, Steffen E and Niederer, Steven A and Rueckert, Daniel and others , journal=. 2024 , publisher=

2024
[19]

Whole heart

Xu, Hao and Muffoletto, Marica and Niederer, Steven A and Williams, Steven E and Williams, Michelle C and Young, Alistair A , booktitle=. Whole heart
[20]

Journal of Cardiovascular Magnetic Resonance , volume=

Automated cardiovascular magnetic resonance image analysis with fully convolutional networks , author=. Journal of Cardiovascular Magnetic Resonance , volume=. 2018 , publisher=

2018
[21]

Nature Medicine , volume=

A population-based phenome-wide association study of cardiac and aortic structure and function , author=. Nature Medicine , volume=. 2020 , publisher=

2020
[22]

Nature Communications , volume=

Genetic basis of right and left ventricular heart shape , author=. Nature Communications , volume=. 2024 , publisher=

2024
[23]

Journal of Cardiac Failure , volume=

Bozkurt, Biykem and Coats, Andrew JS and Tsutsui, Hiroyuki and Abdelhamid, Magdy and Adamopoulos, Stamatis and Albert, Nancy and Anker, Stefan D and Atherton, John and B. Journal of Cardiac Failure , volume=. 2021 , publisher=

2021
[24]

Circulation , volume=

Right versus left ventricular failure: differences, similarities, and interactions , author=. Circulation , volume=. 2014 , publisher=

2014
[25]

UK, Nawwar Al-Attar and Atherton, John James and Bauersachs, Johann and UK, A John Camm and Carerj, Scipione and Ceconi, Claudio and Coca, Antonio and UK, Perry Elliott and Erol,. 2016. European Heart Journal , volume=

2016
[26]

2014 , publisher=

Gupta, Deepak K and Shah, Amil M and Giugliano, Robert P and Ruff, Christian T and Antman, Elliott M and Grip, Laura T and Deenadayalu, Naveen and Hoffman, Elaine and Patel, Indravadan and Shi, Minggao and others , journal=. 2014 , publisher=

2014
[27]

International Journal of Cardiology , volume=

Changes in left atrial size in patients with persistent atrial fibrillation: a prospective echocardiographic study with a 5-year follow-up period , author=. International Journal of Cardiology , volume=. 2005 , publisher=

2005
[28]

Fast implementation of

Sun, Xu and Xu, Weichao , journal=. Fast implementation of. 2014 , publisher=

2014
[29]

Nature Communications , volume=

Environmental and genetic predictors of human cardiovascular ageing , author=. Nature Communications , volume=. 2023 , publisher=

2023
[30]

International Conference on Medical Image Computing and Computer-Assisted Intervention , pages=

Spatio-temporal neural distance fields for conditional generative modeling of the heart , author=. International Conference on Medical Image Computing and Computer-Assisted Intervention , pages=
[31]

2024 , publisher=

Kong, Fanwei and Stocker, Sascha and Choi, Perry S and Ma, Michael and Ennis, Daniel B and Marsden, Alison L , journal=. 2024 , publisher=

2024
[32]

A personalized time-resolved

Qiao, Mengyun and McGurk, Kathryn A and Wang, Shuo and Matthews, Paul M and O’Regan, Declan P and Bai, Wenjia , journal=. A personalized time-resolved. 2025 , publisher=

2025
[33]

Proceedings of the 4th International Conference on Computer Graphics and Interactive Techniques in Australasia and Southeast Asia , pages=

Laplacian mesh optimization , author=. Proceedings of the 4th International Conference on Computer Graphics and Interactive Techniques in Australasia and Southeast Asia , pages=
[34]

2016 , publisher=

Differential geometry of curves and surfaces , author=. 2016 , publisher=

2016
[35]

Mesh-based 3D motion tracking in cardiac

Meng, Qingjie and Bai, Wenjia and Liu, Tianrui and O’regan, Declan P and Rueckert, Daniel , booktitle=. Mesh-based 3D motion tracking in cardiac
[36]

2022 , publisher=

Meng, Qingjie and Qin, Chen and Bai, Wenjia and Liu, Tianrui and De Marvao, Antonio and O’Regan, Declan P and Rueckert, Daniel , journal=. 2022 , publisher=

2022
[37]

Deng, Yu and Xu, Hao and Rodrigo, Sashya and Williams, Steven E and Williams, Michelle C and Niederer, Steven A and Pushparajah, Kuberan and Young, Alistair , booktitle=
[38]

Qiao, Mengyun and Zheng, Jin and Zhang, Weitong and Ma, Qiang and Li, Liu and Kainz, Bernhard and O’Regan, Declan P and Matthews, Paul M and Niederer, Steven and Bai, Wenjia , booktitle=
[39]

2023 , publisher=

Qiao, Mengyun and Wang, Shuo and Qiu, Huaqi and De Marvao, Antonio and O’Regan, Declan P and Rueckert, Daniel and Bai, Wenjia , journal=. 2023 , publisher=

2023
[40]

A bi-ventricular cardiac atlas built from 1000+ high resolution

Bai, Wenjia and Shi, Wenzhe and de Marvao, Antonio and Dawes, Timothy JW and O’Regan, Declan P and Cook, Stuart A and Rueckert, Daniel , journal=. A bi-ventricular cardiac atlas built from 1000+ high resolution. 2015 , publisher=

2015
[41]

Medical Imaging 2001: Image Processing , volume=

Statistical models of appearance for medical image analysis and computer vision , author=. Medical Imaging 2001: Image Processing , volume=

2001
[42]

2008 , publisher=

Statistical models of shape: Optimisation and evaluation , author=. 2008 , publisher=

2008
[43]

Booth, James and Roussos, Anastasios and Zafeiriou, Stefanos and Ponniah, Allan and Dunaway, David , booktitle=. A
[44]

2013 , publisher=

Faghih Roohi, Shahrooz and Aghaeizadeh Zoroofi, Reza , journal=. 2013 , publisher=

2013
[45]

IEEE Journal of Biomedical and Health Informatics , volume=

Statistical shape modeling of the left ventricle: myocardial infarct classification challenge , author=. IEEE Journal of Biomedical and Health Informatics , volume=. 2017 , publisher=

2017
[46]

Nature Machine Intelligence , volume=

Deep-learning cardiac motion analysis for human survival prediction , author=. Nature Machine Intelligence , volume=. 2019 , publisher=

2019
[47]

High-Resolution Maps of Left Atrial Displacements and Strains Estimated with

Galazis, Christoforos and Shepperd, Samuel and Brouwer, Emma JP and Queir. High-Resolution Maps of Left Atrial Displacements and Strains Estimated with. IEEE Transactions on Medical Imaging , year=
[48]

2022 , publisher=

Xia, Yan and Chen, Xiang and Ravikumar, Nishant and Kelly, Christopher and Attar, Rahman and Aung, Nay and Neubauer, Stefan and Petersen, Steffen E and Frangi, Alejandro F , journal=. 2022 , publisher=

2022
[49]

International Conference on Learning Representations , year=

Semi-Supervised Classification with Graph Convolutional Networks , author=. International Conference on Learning Representations , year=
[50]

Auto-encoding Variational

Kingma, Diederik P and Welling, Max , booktitle=. Auto-encoding Variational
[51]

IEEE Transactions on Medical Imaging , volume=

Three-dimensional modeling for functional analysis of cardiac images: A review , author=. IEEE Transactions on Medical Imaging , volume=. 2002 , publisher=

2002
[52]

IEEE Transactions on Medical Imaging , volume=

Automatic construction of multiple-object three-dimensional statistical shape models: Application to cardiac modeling , author=. IEEE Transactions on Medical Imaging , volume=. 2003 , publisher=

2003
[53]

Geometric deep learning: going beyond

Bronstein, Michael M and Bruna, Joan and LeCun, Yann and Szlam, Arthur and Vandergheynst, Pierre , journal=. Geometric deep learning: going beyond. 2017 , publisher=

2017
[54]

Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges

Geometric deep learning: Grids, groups, graphs, geodesics, and gauges , author=. arXiv preprint arXiv:2104.13478 , year=

work page internal anchor Pith review arXiv
[55]

Circulation , volume=

Cardiovascular magnetic resonance , author=. Circulation , volume=. 2010 , publisher=

2010
[56]

Standardized cardiovascular magnetic resonance imaging (

Kramer, Christopher M and Barkhausen, J. Standardized cardiovascular magnetic resonance imaging (. Journal of Cardiovascular Magnetic Resonance , volume=. 2020 , publisher=

2020
[57]

2012 , publisher=

Collins, Rory , journal=. 2012 , publisher=

2012
[58]

2016 , publisher=

Petersen, Steffen E and Matthews, Paul M and Francis, Jane M and Robson, Matthew D and Zemrak, Filip and Boubertakh, Redha and Young, Alistair A and Hudson, Sarah and Weale, Peter and Garratt, Steve and others , journal=. 2016 , publisher=

2016
[59]

Bycroft, Clare and Freeman, Colin and Petkova, Desislava and Band, Gavin and Elliott, Lloyd T and Sharp, Kevin and Motyer, Allan and Vukcevic, Damjan and Delaneau, Olivier and O’Connell, Jared and others , journal=. The. 2018 , publisher=

2018
[60]

2019 , publisher=

Littlejohns, Thomas J and Sudlow, Cathie and Allen, Naomi E and Collins, Rory , journal=. 2019 , publisher=

2019
[61]

2015 , publisher=

Sudlow, Cathie and Gallacher, John and Allen, Naomi and Beral, Valerie and Burton, Paul and Danesh, John and Downey, Paul and Elliott, Paul and Green, Jane and Landray, Martin and others , journal=. 2015 , publisher=

2015
[62]

Linear Algebra and its Applications , volume=

Fast low-rank modifications of the thin singular value decomposition , author=. Linear Algebra and its Applications , volume=. 2006 , publisher=

2006
[63]

IEEE Transactions on Pattern Analysis and Machine Intelligence , volume=

Candid covariance-free incremental principal component analysis , author=. IEEE Transactions on Pattern Analysis and Machine Intelligence , volume=. 2003 , publisher=

2003
[64]

IEEE Transactions on Systems, Man, and Cybernetics , volume=

A novel incremental principal component analysis and its application for face recognition , author=. IEEE Transactions on Systems, Man, and Cybernetics , volume=. 2006 , publisher=

2006
[65]

Wiley Interdisciplinary Reviews: Computational Statistics , volume=

Principal component analysis , author=. Wiley Interdisciplinary Reviews: Computational Statistics , volume=. 2010 , publisher=

2010
[66]

Nature Reviews Methods Primers , volume=

Principal component analysis , author=. Nature Reviews Methods Primers , volume=. 2022 , publisher=

2022
[67]

ACM Computing Surveys , volume=

Principal component analysis: A natural approach to data exploration , author=. ACM Computing Surveys , volume=. 2021 , publisher=

2021
[68]

2023 , publisher=

Wasserthal, Jakob and Breit, Hanns-Christian and Meyer, Manfred T and Pradella, Maurice and Hinck, Daniel and Sauter, Alexander W and Heye, Tobias and Boll, Daniel T and Cyriac, Joshy and Yang, Shan and others , journal=. 2023 , publisher=

2023
[69]

IEEE Transactions on Medical Imaging , volume=

Tobon-Gomez, Catalina and Geers, Arjan J and Peters, Jochen and Weese, J. IEEE Transactions on Medical Imaging , volume=. 2015 , publisher=

2015
[70]

Coronary centerline extraction from

Metz, CT and Schaap, Michiel and Weustink, AC and Mollet, NR and van Walsum, Theo and Niessen, WJ , journal=. Coronary centerline extraction from. 2009 , publisher=

2009
[71]

Medical Image Analysis , volume=

Standardized evaluation framework for evaluating coronary artery stenosis detection, stenosis quantification and lumen segmentation algorithms in computed tomography angiography , author=. Medical Image Analysis , volume=. 2013 , publisher=

2013
[72]

2023 , publisher=

Zeng, An and Wu, Chunbiao and Lin, Guisen and Xie, Wen and Hong, Jin and Huang, Meiping and Zhuang, Jian and Bi, Shanshan and Pan, Dan and Ullah, Najeeb and others , journal=. 2023 , publisher=

2023
[73]

Medical Image Analysis , Year=

Zhuang, Xiahai and Shen, Juan , Title=. Medical Image Analysis , Year=
[74]

Evaluation of algorithms for multi-modality whole heart segmentation: an open-access grand challenge , Journal=

Zhuang, Xiahai and Li, Lei and Payer, Christian and Stern, Darko and Urschler, Martin and Heinrich, Mattias P and Oster, Julien and Wang, Chunliang and Smedby,. Evaluation of algorithms for multi-modality whole heart segmentation: an open-access grand challenge , Journal=. 2019 , Volume=

2019
[75]

IEEE Transactions on Pattern Analysis and Machine Intelligence , Year=

Zhuang, Xiahai , Title=. IEEE Transactions on Pattern Analysis and Machine Intelligence , Year=
[76]

The American Journal of Cardiology , volume=

Two-dimensional echocardiographic methods for assessment of left atrial volume , author=. The American Journal of Cardiology , volume=. 2006 , publisher=

2006
[77]

Journal of cardiovascular magnetic resonance , volume=

Reference left atrial dimensions and volumes by steady state free precession cardiovascular magnetic resonance , author=. Journal of cardiovascular magnetic resonance , volume=. 2010 , publisher=

2010
[78]

2015 , publisher=

SCOT-HEART Investigators, The , journal=. 2015 , publisher=

2015
[79]

Medical Image Analysis , volume=

A deep-learning approach for direct whole-heart mesh reconstruction , author=. Medical Image Analysis , volume=. 2021 , publisher=

2021
[80]

International Conference on Medical Image Computing and Computer-Assisted Intervention , pages=

Whole heart mesh generation for image-based computational simulations by learning free-from deformations , author=. International Conference on Medical Image Computing and Computer-Assisted Intervention , pages=

Showing first 80 references.